VOGONS


First post, by kool kitty89

User metadata
Rank Member
Rank
Member

I put together a Celeron 366 retro rig for a friend a couple months back using a socket 370 Soyo 61ZA-NA (440ZX chipset) motherboard my family had in storage.
At the time of the build, I only bothered to test the BIOS selected 75 and 83 MHz FSP overclocking options as I didn't want to push the system too hard and the board itself had been set to the default 66 MHz jumper (so I assumed my dad hadn't pushed a 100 MHz overclock back when the board was in use some 10+ years ago). I'd gotten the impression that 366=>550 MHz Celeron overclocks were a bit iffy in general, especially without a bigger heatsink and voltage boost to ~2.2V.

However, yesterday I was over at that friend's house and decided to give the 100 MHz jumper a try. To my surprise, with only the stock intel branded heatsink+fan and the board's default 2.08V, it ran just fine in DOS and Win98SE with all the games he had installed (a bunch of Diablo II was the newest thing iirc). Less surprising (especially given the modest voltage and good case cooling) was the CPU temperature remaining quite reasonable as well, peaking just over ~40C/105F. (well within spec for that CPU)
I'd also already been using PC-100 RAM, so that wasn't an issue.

I didn't run any benchmarks yet though, and he didn't have Unreal installed, so I couldn't test the stability there either (I know Unreal tends to be finicky about CPU stability), but otherwise there were no noticeable bugs at all . . . at least beyond some problems that were already present before the overclock.

Reply 2 of 7, by RogueTrip2012

User metadata
Rank Oldbie
Rank
Oldbie

< Jealous.

My first major overclock was a celery 366@550 on a Abit BE6-II. I Still have the parts but it died back in the day and can't revive it 🙁 (yes I tried recapping the boards)

Twas a great rig while it lasted though.

> W98SE . P3 1.4S . 512MB . Q.FX3K . SB Live! . 64GB SSD
>WXP/W8.1 . AMD 960T . 8GB . GTX285 . SB X-Fi . 128GB SSD
> Win XI . i7 12700k . 32GB . GTX1070TI . 512GB NVME

Reply 3 of 7, by luckybob

User metadata
Rank l33t
Rank
l33t

I had a P2-450 running @ 675. It ran for a good 2 months on my asus P3b-f board. Until the cache chips decided to give up and they released their magic smoke. 🙁

It is a mistake to think you can solve any major problems just with potatoes.

Reply 4 of 7, by bestemor

User metadata
Rank Oldbie
Rank
Oldbie

My s370 celery 566@850 (in slot1) has been running 'non-stop' now since 2000, with a rather high(max) voltage*.
(*: don't remember what it actually is now, as I had to do some slotket/bios magic to get back the higher values, and bios value is not correct - it even believes it to be an Piii 803mhz EB, 😜 )

Wonder how long it takes before it dies from this abuse... 🤣

Reply 5 of 7, by kool kitty89

User metadata
Rank Member
Rank
Member
bestemor wrote:

My s370 celery 566@850 (in slot1) has been running 'non-stop' now since 2000, with a rather high(max) voltage*.
(*: don't remember what it actually is now, as I had to do some slotket/bios magic to get back the higher values, and bios value is not correct - it even believes it to be an Piii 803mhz EB, 😜 )

Wonder how long it takes before it dies from this abuse... 🤣

That sort of overclock is pretty common for Coppermine Celerons, especially the lower speed ones (things like 666=>1000 MHz are less common).

Unless the voltage is boosted ridiculously high, it probably isn't hurting it that much either. Even if way out of spec at ~2.0 volts, that's still not crazy for 180 nm parts in general (as with the K6-2/3+) as long as the temp is reasonable.

Such high voltages really shouldn't be needed for that overclock though. Even for the higher overclocks (like 666 to 1000), 1.8V is usually the max needed, and the overclocking guides I've seen list 1.7 V as the upper-end needed for 566 to 850.

Reply 6 of 7, by bestemor

User metadata
Rank Oldbie
Rank
Oldbie

@kool kitty89:
You're probably right. I no longer remember what options I tried during initial set-up (unchanged now for 12+ years!).

Though I do seem to recall that it wouldn't work at that speed for my particular specimen, unless I upped the voltage some more.
I THINK it is at 1.9v now. (either that, or 2.1v)
But as long as it works, right... ? 😎

Reply 7 of 7, by kool kitty89

User metadata
Rank Member
Rank
Member
bestemor wrote:
@kool kitty89: You're probably right. I no longer remember what options I tried during initial set-up (unchanged now for 12+ yea […]
Show full quote

@kool kitty89:
You're probably right. I no longer remember what options I tried during initial set-up (unchanged now for 12+ years!).

Though I do seem to recall that it wouldn't work at that speed for my particular specimen, unless I upped the voltage some more.
I THINK it is at 1.9v now. (either that, or 2.1v)
But as long as it works, right... ? 😎

Cyrix pushed their 180 nm MII parts to 2.2V as the standard setting (and 180 nm K6-2/III+ were specced within 2.1v), so even 2.1V on a Coppermine chip wouldn't be crazy to believe. (as far as the overvoltage not burning up the transistors or interconnect)
Though I'd imagine it runs quite hot.